| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 1908, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.6318477988243103, | |
| "learning_rate": 1.6677148846960168e-05, | |
| "loss": 0.3859, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5580645161290323, | |
| "eval_loss": 0.19784556329250336, | |
| "eval_runtime": 1.3657, | |
| "eval_samples_per_second": 2269.886, | |
| "eval_steps_per_second": 47.594, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.5229398012161255, | |
| "learning_rate": 1.3354297693920338e-05, | |
| "loss": 0.1669, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8090322580645162, | |
| "eval_loss": 0.1015559658408165, | |
| "eval_runtime": 1.5656, | |
| "eval_samples_per_second": 1980.027, | |
| "eval_steps_per_second": 41.517, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.4804494082927704, | |
| "learning_rate": 1.0031446540880504e-05, | |
| "loss": 0.1087, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8664516129032258, | |
| "eval_loss": 0.07101532071828842, | |
| "eval_runtime": 1.397, | |
| "eval_samples_per_second": 2218.976, | |
| "eval_steps_per_second": 46.527, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.43196460604667664, | |
| "learning_rate": 6.708595387840672e-06, | |
| "loss": 0.0856, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8864516129032258, | |
| "eval_loss": 0.057509638369083405, | |
| "eval_runtime": 1.4059, | |
| "eval_samples_per_second": 2205.034, | |
| "eval_steps_per_second": 46.235, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.34330040216445923, | |
| "learning_rate": 3.385744234800839e-06, | |
| "loss": 0.0741, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8916129032258064, | |
| "eval_loss": 0.05096952244639397, | |
| "eval_runtime": 1.5943, | |
| "eval_samples_per_second": 1944.38, | |
| "eval_steps_per_second": 40.769, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.41723889112472534, | |
| "learning_rate": 6.289308176100629e-08, | |
| "loss": 0.0689, | |
| "step": 1902 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 1908, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 1000000000.0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 495439677408900.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.8996260811262161, | |
| "num_train_epochs": 6, | |
| "temperature": 17 | |
| } | |
| } | |